Bash Append header information to each line of a f

Question

I have a file that contains data header information followed by data entries. Each header will be different and I need to add the information from each header onto each line of the file, until the next header get encountered. For instance:

"Header 1","head1_info" date1,data1 date1,data2 "Header 2","head2_info" date3,data5 data4,data6

I want this file to be appended to:

    `"Header 1","head1_info"
    head1_info,date1,data1
    head1_info,date1,data2
    "Header 2","head2_info"
    head2_info,date3,data5
    head2_info,date4,data6`

I've tried assigning grep to variable, but I don't know how to make it change from to the next one, once it encounters "Header 2", I've also been experimenting with sed and awk, but I can't get them to do what I want either. Any guidance would be greatly appreciated.

Answer 1

This might work for you (GNU sed):

sed -r '/^"/h;//!{G;s/(.*)\n.*"(.*)"/\2,\1/}' file

On encountering a header line store it in the hold space. For all other lines, append the hold space to the pattern space and substitute the required info using back references.

Answer 2

Assuming any line which contains a double quote indicates a header line:

awk '/"/{a=$2;gsub("\"","",a); print; next}{print a FS $0}' FS=, input

If there is some other method to determine a header, just change the pattern matched at the start. (This records the value of column 2 from header lines in the variable a. The gsub just strips the double quotes.)